AI that holds up in real operations.
Your org chart isn't where the work happens. It's in the handoffs no one documented, the deck no one showed.
The reframe
What you actually need designed isn't the screen.
When AI goes into your operation, the job isn't shipping an interface. It's getting something that survives first contact with real people and live data — and keeps working after.
That job lives under the screen: what your customer is really there for, what makes them commit, how it reaches them, what it costs you to run. Get the exchange right and the interface almost designs itself.
That's the part I design. The screen is the smallest piece where the value finally shows up.
The scope
Everywhere the value moves.
What your customer is hiring you for.
Before anything gets built, I find the job your customer is actually trying to get done, and where today's experience fails them. That's where the value gets defined, so the system solves the real problem instead of a clever one.
What makes them say yes.
The moment someone decides you're worth it is a design problem. The system is built for that decision, not just the demo before it.
How it reaches the people it's for.
Channels, launch, the loops that grow it: designed in from day one. What works in the room works in the market.
What it costs you to run.
Every exchange has economics. It's designed to the value it creates and the cost to deliver, so the system earns its keep — and the engagement is scoped to what's at stake.
Proof it holds.
Built and measured against your live data, with evals tied to your business rules. If you can't measure the value, you can't trust it.
Snowball Sprint
One motion.
Storyboard
The human in the right place.
Most AI breaks at the seam, not the model. I storyboard the workflow to find where a person stays in the loop and where the AI takes over. That placement decides whether it works.
Digital Twin
Working systems, not wireframes.
I prototype in real code against your real data from day one, so you see how the system actually behaves, and it breaks early, while breaking is still cheap.
Value Matrix
Proven before it's built.
I pressure-test ROI and risk with working prototypes before your engineers write production code, so they build on proven ground, not a guess.
What I need from you.
Access to the people accountable.
Time on the calendar.
The authority to make calls.
The method, applied
Low Ox Life: What Empathy Really Means in Design
Empathy is not a workshop exercise. It is the discipline of building for a problem you have actually carried, and using design to make a hard daily reality navigable for the people living it.
Health & Wellness · 2026
PlanFlow: Structuring How Plan Reviews Get Decided
A plan submission and review system for agencies managing permits and approvals, with structured workflows, role-based reviews, and AI that sharpens human judgment instead of replacing it.
Government & Permitting · 2026
The Bot Was Secure. The System Wasn't.
This site’s own lead bot had 17 security layers and answered every question politely — while running completely unprotected. How a routine audit caught it, and what fail-safe really means.
Applied AI Systems · 2026
Everything I offer.
You leave with a working system.
Not a recommendation deck — running against your real data, with the evals to prove it holds.
Let's talk